Tumors of the head and neck are relatively common epithelial tumors in people. Tumors of the head and neck appear to offer a particularly appealing model for clinical studies since following surgical intervention the rate of local recurrence of tumors is relatively high. In fact it has previously been shown that retinoids can significantly alter the rate of recurrence of tumors of the head and neck. The initial attempts of the Chemoprevention Branch in this specific area entailed the use of the DMBA hamster cheek pouch model. In general we found this model to be unresponsive to a variety of chemopreventives. We wish therefore to examine an alternative model. The first is a model for inducing tumors of the tongue in rats employing the carcinogen 4NQO. As reviewed by Gerson oral lesions in rats produced by 4-NQO are similar to human lesions since manyare ulcerated and endophytic tongue lesions. The objective of the proposed studies will be to investigate a series of four compounds for chemopreventive activity in the 4NQO model and to perform preliminary studies with a single agent in the transgenic model.